Foreclosure Delay and Consumer Credit Performance May 10, 2013 Paul Calem, Julapa Jagtiani & William W. Lang Federal Reserve Bank of Philadelphia The views expressed are those of the authors and do not necessarily reflect the views of the Federal Reserve Bank of Philadelphia or the Federal Reserve System.
Foreclosure Delay and Performance on Non-mortgage Debt Timeline for foreclosures are lengthening Many costs associated with this delay Loss given foreclosure Negative neighborhood externalities One benefit to households in foreclosure is prolonged period of reduced housing expenses Question: Do households take advantage of this benefit by improving their balance sheets? 2
Empirical Results: Impact of Foreclosure Delay on Credit Card Performance We investigate whether longer foreclosure timelines impact a consumer s performance on their credit card loans Where foreclosure timelines are longer, borrowers are more likely to: Cure delinquent credit card accounts Reduce their outstanding balances on credit card accounts Additional Implications: Completion of foreclosure may be an important risk factor for credit card and other consumer loans Credit card delinquencies may rise in the future as backlog of foreclosures are cleared 3
3,000,000 Number of Loans in Foreclosure vs. Foreclosure Started 2,500,000 2,000,000 1,500,000 1,000,000 500,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Foreclosure Started Foreclosure 4
Foreclosure Timelines are Lengthening Cordell and Shenoy (2011) estimate the average number of days from past due to liquidation: Found that timelines lengthened substantially in the last several years Estimates ranged from 16 months in Arizona to 28.9 months in New Jersey Reasons for lengthened timeline Changes in state laws (Gerardi, Lambie-Hanson, and Willen (2011)) Robo-signing moratoria Poor documentation related to the housing boom and use of the Mortgage Electronic Registration System (MERS) (Allen, Peristiani, and Tang (2012)) 5
Percent of Foreclosure Population with Delinquent Credit Card Debt (60+ DPD) 0.5 0.4 0.3 0.2 0.1-8 -6-4 -2 0 2 4 6 8 Quarters Around Foreclosure Start Date Pre-Crisis (2004:Q1-2007:Q3) 6 Crisis (2007:Q4-2011:Q1)
Average Credit Card Balance Around Foreclosure Start Date 12,000 10,000 8,000 6,000 4,000-8 -6-4 -2 0 2 4 6 8 Quarters Around Foreclosure Start Date Pre-Crisis (2004:Q1-2007:Q3) 7 Crisis (2007:Q4-2011:Q1)
Data Sources Data from 2004:Q1 to 2012:Q3 Merged from three sources: Loan-Level LPS Data: Monthly loan level mortgages FRB New York Equifax Consumer Credit Panel: Monthly 5 percent random sample from Equifax consumer credit reporting database FRB Philadelphia Equifax: Semi-annual data set from Equifax. Excludes observations with more than one first-lien mortgage 8
Sample Data Merged data sources by: origination loan amount (exact $ amount), ZIP code, location, and the origination month of the mortgage Results with non-merged data provide qualitatively similar results Includes only individuals: With first-lien mortgages originated in 2000 or later with a foreclosure start date during Q1 2004 to Q4 2010 Not in bankruptcy by the foreclosure start date With at least one seriously delinquent credit card account at foreclosure start date Regression analysis based on sample sizes up to 20,271 observations 9
Variables Used in the Analysis and Hypotheses Dependent Variables Probability of curing at least one seriously delinquent credit card Probability of reducing credit card loan balances Measures of Foreclosure Timeline Average months from foreclosure start to foreclosure end by zip code (+) Loan level months from last day current to foreclosure start (+) Control Variables Change in unemployment (-) County ratio of seriously delinquent credit cards (-) Number of non-mortgage accounts (+) Number of delinquent cards (-) Prior credit card delinquency (-) Credit card utilization (-) FICO Score at Origination (+) Refreshed Risk Score (+) Refreshed LTV (+) Jumbo mortgage dummy variable (+) Subprime dummy variable (-) 10
Empirical Findings Average length of the foreclosure process by zip code is associated with higher probability of credit card cure and reducing balances Loan level months from last day current to foreclosure start is associated with higher probability of cure but is not significantly related to reduced credit card balances Findings are robust to multiple specifications All except one of the control factors has the expected sign Refreshed risk score has a negative sign even though FICO at origination has a positive sign 11
Conclusions Delays in the foreclosure process have significant costs: Higher loss given foreclosure for lenders Possible negative neighborhood externalities For individuals who are being foreclosed upon, longer timelines provide a positive financial benefit through reduced housing expenses Our analysis finds that individuals utilize the benefits of longer foreclosure timelines to improve their performance on credit card debt, rather than increasing consumption Completion of the foreclosure process could be a significant risk factor in the future for credit card lenders 12